## Introduction

### Experiment 2

In the experiments introduced above, we focused on positive and negative varieties of reinforcement as a model for earning and saving. Given the acquired negative valence and disutility of the saving condition, we would have expected greater attention11,18,35, yet we found a savings posteriority effect. Casting a condition as “savings” inverted what we would be expected from its equal arousal and increased negative value relative to earning. Thus, saving has connotations that counteract its negative value to be less salient.

### Data analysis

For details, see the Supplementary Information. Performance in the value reinforcement task was measured by participants’ RTs on correct trials, error rates, and proportions of gains received and successful losses avoided. The significance level for the main statistical analyses was set at p < 0.05. Holm-Bonferroni correction was applied to the alpha criterion for multiple comparisons when determining significance. To assess the TOJ task, we calculated each participant’s PSS using a Gaussian fit method. We also analyzed participants’ accuracy (i.e., mean percent of correct responses) submitting hit rates to an ANOVA (temporal order of the presenting stimuli × SOA (8, 18, 38, 68, and 98 ms)52. Following Cohen69, effect sizes were reported as η2 (small, 0.01; medium, 0.06; large, 0.14) for ANOVAs, and as d (small = 0.20; medium = 0.50; large = 0.80) for planned comparison t tests.

### Data availability

The data that support the findings of this study are available from the corresponding authors upon request.